基于行为的应急交通疏散建模与仿真研究

项目名称： 基于行为的应急交通疏散建模与仿真研究

项目编号： No.61203166

项目类型： 青年科学基金项目

立项/批准年度： 2013

项目学科： 自动化学科

项目作者： 吕宜生

作者单位： 中国科学院自动化研究所

项目金额： 24万元

中文摘要： 应急交通疏散是处置人为或自然灾害事件，防止和减少人员生命财产损失的有效手段。由于无法或很难对应急交通疏散进行现场实验，且突发事件下人员的行为对整个应急交通疏散管理与控制有重要的影响，本项目将从突发事件下人员的个体和家庭决策行为角度切入，利用概率图模型研究人员在突发事件下的活动链，采用Multi-Agent建模方法，构建集成整个疏散过程中人员行为和交通理论的应急交通疏散模拟仿真系统，探索不同的交通行为模式对应急交通疏散效率的影响，尝试从行为诱导与控制角度给出基于行为的应急交通疏散控制新方法。该研究将有助于人们更好的认识应急交通疏散过程，揭示突发事件下人员的行为对应急交通疏散效率的影响，提供一种应急交通疏散控制新方法，能为政府管理部门提供辅助决策支持，具有十分重要的理论和实际应用价值。

中文关键词： 应急疏散；交通仿真；概率图模型；计算实验；仿真优化

英文摘要： Emergency Traffic Evacuation is an effective response and mitigation strategy to protect human life and property loss against man-made or natural disasters. It is very difficult, many times even infeasible to conduct field experiments on emergency traffic evacuation. However, it is necessary to consider evacuee behavior under emergency conditions in traffic evacuation because evacuee behavior has a significant impact on emergency traffic evacuation mangement and control. In this project, we will cut into emergency traffic evacuation modeling and simulation from the perspective of personal and family decision behavior. We will firstly study human activity chain under emergency situations with propabilistic graphical model, and then build an Agent-based microscopic emergency traffic simulation system integrating human behavior and traffic theory during the whole evacuation process. Next we will explore the impact of different traffic behavior patterns on emergency traffic evacuation efficiency, and try to provide a new kind of behavior-based emgergency evacuation control methods from the perspective of behavior guidance and control. This study will help to learn more about emergency traffic evacuation process, explain the impact of human behavior on emergency traffic evacuation efficiency, provide a new kind of em

英文关键词： Emergency Evacuation；Traffic Simulation；Probabilistic Graphical Model；Computational Experiments；Simulation Optimization